Download This PDF is a selection from an out-of-print volume from... Bureau of Economic Research

This PDF is a selection from an out-of-print volume from the National
Bureau of Economic Research
Volume Title: Foreign Trade Regimes and Economic Development: Chile
Volume Author/Editor: Jere H. Behrman
Volume Publisher: NBER
Volume ISBN: 0-87014-508-8
Volume URL: http://www.nber.org/books/behr76-1
Publication Date: 1976
Chapter Title: The Foreign Sector and Chilean Economic Development:
An Overview
Chapter Author: Jere H. Behrman
Chapter URL: http://www.nber.org/chapters/c4024
Chapter pages in book: (p. 3 - 45)
I
Chapter
1
The Foreign Sector and Chilean
Economic Development:
An Overview
1.1
INTRODUCTION
In discussions of economic development experience, considerable emphasis has
been placed on the foreign sector. This emphasis partly reflects the view that
the foreign sector can play a key role in the development process through the
exploitation of dynamic comparative advantage, the transfer of technology and
of command over resources, and inducements for efficiency,' and partly the
fact that, in developing economies, the foreign sector is particularly subject to
government influence.
Analyses of the foreign sector in actual economic development experience, however, usually have had definite limitations. In many cases the importance of quantitative restrictions in the foreign sector regime has been disregarded or understated. Historical perspective often has been lacking. Too
frequently the supporting evidence is impressionistic, and available quantitative information has been ignored or not subjected to rigorous analysis. Repeatedly other macroeconomic policies and the problems to which they relate
have not been kept in their proper perspective, with the result that foreignsector considerations may be, or may appear to be, overstressed. At times
partial-equilibrium conclusions have been misleading or wrong. In some cases
very strong assumptions have been made about rigidities.2 Considerable disagreement therefore remains about the impact of the foreign sector on developing economies.
In this study, I attempt to contribute to the analysis of the relations be3
r
4
INTRODUCTION
tween international economic policies and various macroeconomic goals by a
thorough examination of the experience of one specific developing country—
Chile. Included among the macroeconomic goals are not only growth, but also
other factors that are related to a broader definition of economic development:
external position and the degree of national autonomy, stability in real and
nominal terms, resource allocation and structural change, and the distribution
of control over income and other resources. Some of these factors may be
complementary and others may be competitive with the goal of growth.
Chile is a particularly interesting country for such a study for several
reasons: (i) It has experienced a wide range of foreign-sector policies. (ii)
Since the Great Depression, its attempts at industrial development under
foreign-exchange control and trade restrictions have shown characteristics
common to a number of other developing countries: a large and heavily protected industrial sector, recurrent balance-of-payment crises, a very large for-
eign debt, and high inflation which has helped to perpetuate overevaluation of
the currency. (iii) Policy in the foreign sector has included responses to both
substantial external and internal stimuli. (iv) Chile has had three major liberalization episodes in the past two decades, which provide a rich base for the
evaluation of liberalization efforts in developing countries. Cv) Recent Chilean
governments have varied considerably in their political orientation. (vi) The
Chilean experience has served as one of the major bases for the development
of critiques of the conventional wisdom concerning development. To cite one
important example, the "structuralist school" analysis of inflation and stagnation was based considerably on Chilean experience and was developed in substantial part by economists in Chile. According to the structuralists, the foreign
sector is an important element in the vulnerability of certain Latin American
economies to inflation and stagnation because slow and unstable export growth
1.
B
ti
S1
fa
er
Ifl
te
ei
ti
P
a
a
C
I
1
and a secular deterioration in the terms of trade for exports of primary
l
in government revenues from foreign trade taxes.3
Ii
products result in inadequate import capabilities and procyclical fluctuations
The interactions between the foreign sector and the rest of the Chilean
economy have been quite complicated and have varied substantially over time.
Moreover, in Chile more than in other countries, past events seem to have had
unusually strong effects on recent policy decisions because of the catastrophic
impact of some of those events, especially the Great Depression, and due to
the unusually strong influence of Chilean economists who have attempted to
understand the present in terms of the past.4 In the next three sections of this
chapter, I provide a chronological overview of the Chilean experience, in order
to place the foreign sector in perspective both in a historical sense and as it
relates to other areas of macroeconomic concern. In the last section, I furnish
an outline of the study and a summary of my conclusions,
e
a
U
F
a
a
AN OVERVIEW
a
era1
Ider
tics
fori of
oth
the
ean
The
aent
one
ma-
ub-
ign
can
wth
ary
DflS
ne.
iad
hic
to
to
iüs
der
it
ish
1.2
5
THE PERIOD BEFORE THE
GREAT DEPRESSION5
Between the Spanish discovery of the land area that is now Chile, in 1535,
and the War of Independence starting in 1810, Spain imposed a very mercantiistic regime on Chile. Trade could be conducted only with Spain and the
Spanish colonies; goods were required to be carried in Spanish boats; manufacturing in Chile was discouraged, and the role of the mother country in government was substantial (although the geographic isolation of Chile resulted
in considerable local autonomy). In the sixteenth century, exports consisted
primarily of gold and silver; leather and animal fat were added in the seventeenth century; and copper and wheat were added in the eighteenth. From the
beginning, then, mining products were an important component of Chilean
exports and accounted for over half of the value of exports by the end of the
eighteenth century. At that time imports consisted primarily of sugar (24 per
cent of the total) and tobacco (24 per cent) from Peru, and textiles (16 per
cent) and Paraguayan tea (24 per cent) from Argentina. Thus, Chile had
relatively more of her trade with neighboring Latin American areas at that
time than she was to have in the next 170 years. At the end of the colonial
period, the population of about I million was approximately the same in size
as when the Spanish first arrived, but its composition had changed from being
almost entirely Indian to approximately 80 per cent Spanish and Creole. In
comparison with most other Latin American countries, this population was
relatively homogeneous and not dependent on agriculture. Furthermore, Chile
had a relatively well-developed transportation system and urban structure.
From 1810 until 1830 the Wars of Independence and the subsequent po-
litical consolidation dominated events in Chile with substantial negative effects
on the economy. By 1830 under the leadership of Diego Portales, banditry had
been reduced considerably, public administration had been reorganized, public
financing had been strengthened, and public works expenditure had increased.
Chile then entered into a period of political stability which was unique in mid-
nineteenth century Latin America. The period from 1830 to 1860 is widely
thought to have been a "golden era" of great expansion and development.
Over the long term, international economic policy had moved from the
extreme restrictiveness of colonial mercantilism, in 1810, through phases II
and HI and, by 1850—60, into Phase IV.° The most important change in the
use of quantitative policy tools took place with the enactment of the Law of
Free Commerce of 1811, which opened the ports of Talcahuano, Valparalso,
and Coquimbo to all trading nations. In 1849, the remaining ports were opened,
and the prohibition on the participation-of foreign boats in coastal trade was
p
6
INTRODUCTION
TABLE 1.1
Exchange Rate and Selected Series on Exports, Imports, and Government Income
and Expenditures of Chile, 1820—1930
Exchange
Rate
(pesos,
or
escudos,.
Year
per U.S.
do!.)
(1)
Aver. Annual Exponential
Growth Rate in Past 5 Yrs.
(U.S. dol.)
Imports Exports
(2)
Total
Govt.
Income
(3)
(4)
1820
1825
Ratios to
Total Govt. Income:
Foreign Trade Taxes
Total
(5)
Imports
Exports
(6)
(7)
182
.44
.46
182
1830
1835
1840
1845
1.31
1850
1855
1860
1865
1.06
.091
.138
.106
.57
.62
.52
.57
1.09
.084
.08!
.070
.56
1.14
.028
.046
.046
.56
0.70
.089
.100
.230
.23
.19
.04
1870
1.08
—.030
—.077
—.056
.33
.31
.02
.008
.078
.36
.01
1.95
.045
.002
—.044
.37
1.61
.053
—.118
.019
—.063
2.08
.086
.045
.067
2.93
—.064
—.053
.085
1900
2.97
.122
.162
.02!
.20
.37
.26
.17
.04
1890
1895
.24
.60
.65
.52
.59
.21
.38
1905
3.19
.091
.108
.103
.41
.26
.15
1910
1915
4.62
6.07
.093
—.134
.028
—.000
.030
.19
.13
.31
—.078
:50
.52
1920
1925
1930
5.73
8.55
8.26
.152
.036
.033
.110
.015
.087
.40
.12
.28
.086
.36
.29
.15
.21
1875
1880
1885
1.29
1.29
1.29
1.13
.074.
.048
.101
—.063
.081
Yea
.50
183
183
184
184
185
185
186
.19
.23
186
187
187
181
181
.39
.35
18
1
.39
.10
19
19:
19.
(continued)
rici
Dc
II
7
AN OVERVIEW
TABLE 1.1 (concluded)
Ratios to
Total Govt.
Income:
Extraordinary
Incomea
Development
Functionsb
Total
(8)
(9)
(10)
.
Exports
Year
.04
.02
.01
.04
.23
.39
'.35
.31
.39
.28
.21
,.10
Ratios to
Total 0ovt. Expenditures
1820
1825
1830
.00
1835
.14
Debt Service
.
.
Foreign
(11)
Ratios of External
Public Debt to
Ordinary
Govt.
Income
Export
Value
Aver.
Actual
Import
Tax
Rates
(12)
(13)
(14)
2.7
1.16
.19C
1.8
0.42
.22C
.20C
.
1840
1845
1850
1855
1860
1865
.00
3.1
2.5
.06
.06
.05
.15
.15
.12
.13
.06
.10
.08
1.1
0.35
.14
.16
.13
1.9
0.56
.22c
.56
.32
.10
1.5
0.44
.15
1870
1875
1880
1885
.40
.24
.17
2.5
1.07
.21
.24
.22
.16
2.7
1.22
.20
.36
.16
.10
1.9
1.05
.17
.09
.48
.35
1.5
1.08
.36
1890
1895
1900
.10
.34
.14
.44
.16
.07
.10
1.8
.15
2.3
2.2
1.38
.23
.36
.28
.26
2.66
1.48
.22
.32
.24
1905
.46
.28
.13
.13
2.0
0.88
.31
1910
1915
.49
.00
.33
.31
.13
.18
.13
.17
1.9
2.5
1.03
1.35
.16
1920
1925
1930
.02
.26
.47
.11
.09
.16
.07
1.0
1.4
1.9
0.49
0.69
.10
.16
.26
.04
.31
.31
.21
.19
.08
1.7
1.86
.14
SOURCE: Calculated from data in Humud [1969:18, 100, 110, 130, 173, 205, 207,
and 256].
a. Extraordinary income is from internal or external borrowing.
b. Includes expenditures by ministries of Industry, Public Works and Railroads, Agriculture and Industry, Land and Colonization, Commerce and Communications, and
Development.
c. Overstates import tax rate since numerator includes import and export tax rev-
enues.
I
r
8
INTRODUCTION
suspended. Price-related policy tools were limited primarily to import and cxport taxes, with ad valorem—equivalent explicit rates of no more than 30 per
cent. A unified and almost constant nominal exchange rate (NER)7 (with
some appreciation against the dollar) prevailed from 1830 to 1860 (column
qi
II
1 in Table 1.1), during which time the value of the escudo-equivalent in 1960
United States dollars averaged 1,529 and never varied by more than 5 per cent
from that average (a degree of stability not again attained)
•
•
•
The dating of the switch from Phase IV to Phase V is somewhat arbitrary.9 Domestic support of laissez faire policies was growing as the secular
liberalization in foreign trade progressed. Prominent in both developments was
a fervent French laissez faire advocate, Jean Gustave Courcelle-Seneuil, who
was an adviser to the minister of finance and a professor at the University of
Chile from 1855 through 1863. Courcelle-Seneuil directly participated in the
drafting of important economic legislation during his years in Chile and is said
to have had considerable indirect impact through the dogmatic application of
his theories by Chilean disciples over the next fifty years. Among the pieces
of legislation which he drafted, the 1860 banking law, which in the words of'
Hirschman [1963:164], "reflected distrust of the government" by establishing
"the principle of free, almost wildcat, banking," has become at the least an
important symbol of the acceptance of laissez faire by the Chilean body politic.
Since the 1857—6 1 financial crisis, which reputedly originated in fluctuations in
international markets and led, in some degree, to passage of the 1860 banking
law, also resulted in a break with the past, 1861 has been selected as the year
of transition to a new phase.
In the I 860s and on into the late 1 870s, a very liberal international
economic policy prevailed, except for the suspension of convertibility during
the war with Spain (1865 to mid-1866). During this phase Chilean currency
first appreciated against the dollar (partly because of the Civil War in the
United States) and then started a long secular depreciation (Table 1.1, column
1). The average value of an escudo-equivalent in 1960 United States dollars,
however, remained at about the average 1830—60 level (i.e., $1,512 for 1861—
77 as compared to $1,529 for 1830—60).
A succession of events occurred between 1878 and 1882 which had a
strong impact on the Chilean economy for the next half-century. First, convert-
ibility was suspended in 1878 during a financial crisis largely brought on by a decrease of 27 per cent in the real value of exports between 1872 and 1878. This
decrease, in turn, was primarily due to a succession of bad harvests and a sharp
decline in copper prices. Hirschman [1963: 169] probably is correct in hypoth-
esizing that the 1878 financial crisis by itself probably would not have had
substantially more impact on longer-run Chilean economic events and policies
than did the temporary suspension of convertibility during the 1865—66 war
with Spain had not a constellation of subsequent significant events followed so
11
dl
b
V
5'
a
a
C
5
e
AN OVERVIEW
9
quickly. But other events did follow, and Chile never again returned to the
L
L
pho
of
rId
r0f
gold standard for any period longer than six years.
Second, partially in response to the threat of having her rather substantial
interests in the nitrate mines in the Atacama desert displaced, Chile fought the
War of the Pacific with Peru and Bolivia from 1879 through 1882. Hope of
an immediate return to convertibility was dashed by the war, and temporary
restrictions on imports of nonwar materials were imposed which moved foreign
trade policy back to Phase II or III. Such results were rather minor, however,
in comparison with the longer-run impact of the war. Chile won and incorpo—
rated into her area Tarapacá and Antofagasta, which contained the world's
only known sources of natural sodium nitrate. A new era began as nitrates
dominated Chilean exports, becoming the major source of government revenue
for the next half-century; and Chile became very integrated into the world
economy.
Third, in 1882 the Chilean government denationalized the nitrate industry
by returning ownership to the holders of certificates originally issued before the
War of the Pacific by the Peruvian government.
Fourth, in 1882, after three centuries of wars, the Indian resistance in the
south finally was crushed. The settling of this frontier together with the northem province won in the War of the Pacific led to the creation of new landholding and mining classes which were to join with the middle class to form
a significant political power in the early twentieth century.
For the next half-century following the turbulent war years of 1878—82,
a Phase IV or V international regime generally prevailed. Events and policies
that resulted in some significant alterations in the foreign sector included (i)
attempts to return to complete convertibility, (ii) the tariff revisions of 1897
and 1928, (iii) foreign debt accumulated in the War of the Pacific and the Civil
War of 1891 and in the expanded public works programs of the 1920s, (iv)
changes in world markets, (v) the resurgence of copper exports in the early
twentieth century, (vi) and the establishment of the Central Bank in 1925.
For this half-century preceding the Great Depression, nevertheless, foreigna
sector policies were relatively homogeneous and predominantly liberal.
Five questions about that half-century of experience merit further con-
sideration: (i) Why was complete and enduring convertibility not re-estab-
lished after the War of the Pacific? (ii) Why did ownership of the major cx-
port commodities shift from Chilean to foreign hands? (iii) What was the
evolution and impact of import-substitution policies? (iv) Did increased foreign dependence and vulnerability result from liberalization? (v) What was
the relationship between liberalization and growth?
0
•
10
1.2.1
INTRODUCTION
Failure to Re-establish Enduring Complete Convertibility.
On the face of it, the acquisition of control over the world's only nitrate
source would seem to have made a return to convertibility rather easy after the
War of the Pacific, given the critical role of nitrates in the manufacture of
explosives and fertilizers. In the immediate postwar period, however, convertibility was not re-established because, for the reasons described below, there
was a shortage of foreign exchange at the previous exchange rate level:
i. In the first half of the 1880s total export value actually declined (see
column 3 in Table 1.1) because the increase in nitrate exports was more than
offset by decreases in exports of copper and wheat.
ii. At the same time, government debt service on loans from abroad
reached levels greater than one-third of total government expenditure, in part
because of obligations incurred during the war (Table 1.1, column 11).
iii. Because of the 1882 decision to return nitrate ownership to certificate
owners and because of earlier speculation in such certificates by foreigners, the
Chilean share in nitrate ownership increased only from 20 per cent in 1875 to
36 per cent in 1882. Because the foreign speculation had been funded largely
by the Chilean banking system, moreover, no substantial capital inflow accompanied the assumption of control of almost two-thirds of the nitrates by foreign interests.
iv. Because of internally financed government deficits and wildcat banking, the rate of inflation increased to a level of about 5 per cent per year at the
same time that worldwide many prices were failing.'0
Attempts to return to a metallic standard were made in 1887, 1895—98,
and 1925—31. But all of these attempts were abortive because of a combination
of factors, including the continuation of internal inflation, the setting of fixed
NERs at overvalued levels, the large foreign debt incurred in the Civil War of
1891, and declines in external markets. Between 1878 and 1927 the value of
Chilean currency depreciated steadily (Table 1.1, column 1). The escudoequivalent in 1960 U.S. dollars decreased at a mean rate of 3.8 per cent per
year with the result that in 1927 it was worth only approximately 15 per cent
of its 1878 level.
1.2.2
Shift to Foreign Ownership of Major Exports.
Before the War of the Pacific the major exports were copper and wheat,
both of which were almost entirely domestically owned. For three decades
after the war, copper production fell steadily to a low in 1911 of less than 40
per cent of the 1876 peak level, and the Chilean share of world copper production fell from 44 per cent in 1878 to less than 5 per cent for 1900—15."
Agricultural exports (primarily wheat) also declined, from about 45 per cent
AN OVERVIEW
f
ee
an
ad
te
he
to
lv
r
11
of the total in 1844—80 (Pinto [1962]) to approximately 10 per cent in the
1920s.1' Nitrates became the dominant export in the postwar decades,'3 but
ownership was largely in foreign hands because of the 1882 denationalization
decision. After 1910 there was a resurgence of copper production, but it occurred almost entirely in foreign-owned mines.'4 Production increased at a
mean annual exponential rate of 11.8 per cent from 1910 through 1927; the
Chilean share of world production increased from 4 to 16 per cent in the same
period; and copper again became the leading Chilean export. Thus the weight
in exports shifted from domestically owned copper and wheat to largely foreign-owned nitrates and then to almost completely foreign-owned copper.
This move to foreign control over exports reflected a number of factors:
i. Because of transportation costs, Chile could not compete effectively
with new low-cost wheat producers in North America, Russia, and Argentina.
ii. The richest copper deposits had been exhausted by 1880, resulting in
a substantial increase in Chilean copper-mining costs.
iii. Overvaluation of the currency after the War of the Pacific discouraged
exportation of copper and wheat, the two previously dominant and domestically controlled export commodities.
iv. A shortage of government funds, arising from heavy service charges
on war debts, was an important consideration in the 1882 denationalization
decision.
v. In the early twentieth century, the lack of sufficient funds from domestic sources and the limited access to the international capital market precluded Chilean-controlled application of the new capital-intensive Jaciding
process to the remaining relatively low-grade copper ore.
vi. Before the discovery of a practical method for fixing nitrogen from
the air, in 1914, Chile's nitrates were too important a war import material for
the major powers to let control rest with Chile.
vii. An economic "open door" policy was thought to be useful in attracting desired immigrants.
viii. The philosophy of the ruling elite generally was laissez faire—par-
tially because of the continuing influence of disciples of Courcelle-Seneuil. This
viewpoint was instrumental in the 1882 decision to denationalize nitrates (so
that the government would not be involved in direct production) and in subse-
quent decisions to leave the mining industry almost completely unregulated
except for the collection of export taxes on nitrates.1'
ix. The ruling and commercial elite apparently identified their interests
strongly with those of foreign economic entities and with foreign trade. As a
result, and despite the general laissez faire policies toward the export minerals,
Humud [1969:38, 40, 217] notes, the government did attempt to aid the
foreign-controlled nitrate companies in times of labor-market or internationalmarket crises."
I
12
INTRODUCTION
Because of the shift of the export minerals to foreign ownership, large
transfers of funds were made from Chile to the more developed countries. In
the early twentieth century, there were also substantial capital inflows, which
were used in developing foreign-owned, large-scale copper mines. The evaluation of these net flows and of other costs and benefits is very difficult because
the options cannot be determined with confidence on the basis of the scanty
information available. However, from the nationalistic viewpoint of the 1 960s
and 1970s, which is not the same as that of the ruling elite of those times, it
appears that Chile could have gained much more than she did from her mineral riches by exercising greater control, if not actual ownership.
1.2.3
Evolution and Impact of Import-Substitution Policies.
Conflicting claims about the pattern of import substitution before the
Great Depression abound in the Chilean literature. The following data may
aid in determining the actual evolution and impact of import-substitution
policies:
i. From 1830 to 1960, the ad-valorem-equivalent import tax for final
consumer goods ranged from 20 to 35 per cent; for intermediate goods, from
10 to 15 per cent; and for capital goods, zero (Humud [1969:122, 125]).
Thus, the tariff structure resulted in higher effective than nominal protection
rates for consumer and intermediate goods industries. Export taxes included an
element of protection for the processing of wheat in that the rate for flour was
two-thirds that for grain. In the early part of the period from 1861 to 1896,
some rationalization of the import tariff structure was made by reducing rate
variances between categories. The highest ad-valorem-equivalent nominal rate
also was decreased from 35 per cent to 25 per cent; so some decline in effective
protection for final consumer goods resulted. Under Law 980 of 1897 the
maximum import tariff rate was increased to 60 per cent for some final consumer goods and reduced to zero for some intermediate goods. Effective protection on final goods therefore increased substantially, although effective
protection on some intermediate goods decreased.18
ii. Calculations based on data in Lagos [1966:26} indicate that the share
of consumer goods in imports dropped from 89.6 per cent in 1870—72, to 76.2
per cent in 1878—82, and to 54.8 per cent in 1903—07. This secular decline in
the relative importance of consumer goods imports is consistent with the claim
of increasing substitution for such imports.
iii. The average actual import tax rates (Table 1.1, column 14) do not
indicate a substantial decline immediately after 1860, but do show an increase
from 1885 to 1905, a decline from 1910 to 1925, and an increase as of 1930;
but these rates must be interpreted with some care because of changes in the
composition of imports.
.9
L
13
AN OVERVIEW
iv. Mufloz [1968:82] estimates that the total import substitution coefficient for the period from 1914—15 to 1927 was 0.43 1, as compared to a "norma!" Chenery value of 0.194.19
v. For 1914—15 through 1929—30 Muñoz [1968:38] gives a mean annual
rate of growth for real industrial GDP of 4.8 per cent. Other estimates range
from 2.2 per cent to 6.0 per cent (see Table A.1, line 2.1; see also Ballesteros
and Davis [1963] and CEPAL, in Muñoz [1968:38]).
vi. Both Lagos [1966:144] and Mamalakis [197 lb:306] maintain that
by the 1920s over a fifth of the total labor force was employed in manufacturing. For this sector to have been such a large employer, considerable industrialization must have taken place in the preceding seventy or eighty years.
Thus, the evidence suggests that at least moderate protection of final consumer goods had prevailed since the 1830s.2° The degree of protection was reduced slightly in the 1 860s, and a significant increase was instituted by the
P
tariff law of 1897. The years from 1860 to the Great Depression were not
characterized by substantially less protection than had prevailed before 1860,
however, as Jobet [1955:42], Lagos [1966:19, 22—23, 38], Pinto [1962:16,
21, 49], and Septulveda [1959:36, 71—72] contend; nor did relatively low levels
of protection prevail until just before the Great Depression, as Ellsworth
[1945:49] and Jeanneret [1971: 145] suggest. Moreover, industrial growth
S
from 1860 to 1930 must have been substantial, although not extraordinary, in
order for a manufacturing sector to have developed as large as that which
existed in the 1920s. Much of this development took place, finally, through
import substitution for consumer goods — a process which occurred at more
than the Chenery "normal" rates, at least beginning with the outbreak of World
War I and continuing thereafter.
1.2.4
Liberalization, Foreign Dependence, and Vulnerability.
The contention of the structuralist school (see section 1.1) is that because
of liberalization Chile became extremely dependent on the foreign sector and,
therefore, more vulnerable to instabilities abroad than would otherwise have
been the case. The structuralist claim of great dependence is based on the
following considerations:
i. in the years immediately preceding the Great Depression, the economy
was very dependent on exports and imports. COMMENT: This hypothesis is
based on observation of ratios of exports and of imports to GDP. Available
estimates of these ratios range from 0.3 to 0.6, which are quite high compared
with subsequent Chilean experience and with the experience of most other
countries.21
•
ii. As liberalization took place, the rates of growth of real exports and
imports increased to relatively high levels. COMMENT: A crude test of this hy-
a
14
INTRODUCTION
pothesis is provided by the following two regressions, in which the natural
logarithms of the value of exports (X) and of imports (M), both in 1960 dollars, are the dependent variables and dummy variables are utilized to test
whether significant phase-related deviations in growth rates occurred in the
period 1844—1927: 22
ln(X) = 0.037 TIME + 0.014 TJME*DUfyJ(44_60) + 6.77 (1.1)
(7.2)
(1.1)
= 0.94, SE = 0.23, DW =
ln(M) =
0.048
(5.7)
2.1,
= 0.79
(24.9)
(1.2)
TIME+0.019
(1.6)
—0.012 TIME.DUM(78—27) + 6.49
(2.1)
= 0.93, SE = 0.22, DW =
(27.2)
1.7, , = 0.73
where
TIME = time trend;
DUM(44—60) = dummy variable with value of 1.0 for 1844—60
and 0.0 for other years;
DUM(78—27) = dummy variable with value of 1.0 for 1878—1927
and 0.0 for other years.
These estimates suggest that real exports and imports in fact did not increase
more rapidly in the more liberal, post-1860 phases. To the contrary, the growth
rates were, if anything, significantly higher during the period before 1860. For
exports, such a result might seem surprising because of the acquisition of a
major new export — nitrates — in the War of the Pacific. But apparently the
decline of wheat and copper (mentioned above) offset the expansion of
nitrates. Finally, the implied growth rates for exports, 0.051 for 1844—60 and
0.037 for 1860—1927, and for imports, 0.067 for 1844—60, 0.048 for 1861—
77, and 0.036 for 1878—1927, are less than those reported for some of the
more restrictive, post-1940 phases (lines 1.2.4.1 and 1.2.4.2 in Table A.!),
and undermine the assertion that exports and imports increased quite rapidly
from
1860 to 1930.
the period before the Great Depression, government income was
dependent on the foreign sector. COMMENT: By any available index the
reliance of government income on the foreign sector was considerable: between
1820 and 1930, foreign trade taxes ranged from 23 per cent to 65 per cent of
total government income; for 1908—27, taxes related to foreign income (including income taxes on copper mining) averaged 83 per cent of total taxes;
between 1820 and 1930, extraordinary income (including substantial foreign
iii. In
very
15
AN OVERVIEW
borrowing) was as high as 56 per cent of total government income; between
1840 and 1930, the resulting external public debt service was as high as 35
per cent of total government expenditures; in 1830—1930, the ratio of foreign
public debt to ordinary annual government income was never below 1.0; and
from 1845 to 1930, the ratio of external public debt to annual export value
often exceeded 1.0 (Table 1.1, columns 5, 8, 11, 12, and 13, and Table A.1,
lines 1.2.6.5 and 1.2.6.6). Regressions of ordinary (primarily taxes) and
extraordinary (internal and external borrowing) government income on imports and exports for 1844—1927 provide further confirmation:
)
OGY=0.11M+ 0.40X+371.7
(1.7)
(11.7)
= 0.96, SE = 703., DW =
EGY
(1.0)
(1.3)
1.8, , = 0.71
0.33M — 0.1OX + 293.9
(2.9)
(1.4)
(0.6)
= 0.53, SE = 1,342., DW = 2.1,
0.58
(1.4)
where
OGY = ordinary government income (in thousands of 1960 dollars);
EGY = extraordinary government income (in thousands of 1960 dollars).
•
•
•
I
Variations in the foreign trade variables were consistent with a substantial proportion of the variance in ordinary government income and with over half of
the variance in extraordinary government income.
Relation 1.4 merits further interpretation. The positive coefficient for the
value of imports probably reflects both a secular trend in the use of extraordinary income24 and the dependence of marginal imports on foreign exchange
obtained from external extraordinary government income. The negative coefficient for the value of exports, although significantly nonzero only at the 10
per cent level, provides some support for the claim that the government sub-
stituted between extraordinary income sources and ordinary export tax revenues. When the latter declined the former were increased. To the extent that
such extraordinary sources consisted of external debts (which were more important than internal ones), a decline in the value of exports resulted in an
increase in foreign debt and in foreign dependence.
Although the foregoing evidence indicates that government income depended substantially on the foreign sector, it does not follow that this depen-
dence was unavoidable. In fact, Humud [1969:151] reports that before the
establishment of an income tax, in 1925, internal taxes generally declined as
a percentage of total taxes as a result of deliberate decisions to eliminate most
of these taxes.25
16
INTRODUCTION
iv. With increased liberalization, government revenues became more dependent on the foreign sector. COMMENT: A test of this hypothesis is to inelude phase-coincident dummy variables in relations 1.3 and 1.4. In neither
case, however, does such a procedure result in coefficients that are significantly
nonzero even at the 15 per cent level.26 Examination of the data in Table 1.1
also indicates the lack of associatiOn between dependence on the foreign sector
for government revenues and phases of restrictionism. Foreign trade taxes accounted for over half of total government income both in the relatively restrictive years from 1830 to 1860 and in the more liberal and highly-nitrate-
U
.
S
dependent period from 1885 to 1915 (columns 5, 6, and 7). The external
public debt was relatively large primarily after wars and moderately so after
borrowing abroad for infrastructural development (Table 1.1, columns 12
and 13).27
v. With increased liberalization, foreign ownership increased. COMMENT:
As is discussed above, after the War of the Pacific,28 foreign direct ownership
of the key mineral exports increased considerably.
Thus the available evidence suggests that dependence on the foreign seetor was quite substantial before the Great Depression, although not necessarily
correlated with the degree of liberalization. The assertion that the economy became more vulnerable to externally generated fluctuations is more difficult to
explore, but the following observations have some bearing on this issue.
i. Deviations from the secular time trend provide a measure of fluctu-
ations, some of which, of course, originate in internal events and wars or reflect
government decisions. The mean absolute proportional deviations of actual
export value levels relative to the levels predicted by equation 1.1 are 0.10
(standard deviation of 0.06) for 1844—60, 0.11 (0.12) for 1861—77, and
0.19 (0.24) for 1878—1927. The mean for the last period is significantly
larger at the 5 per cent level than those for the two earlier periods.29 The implication of these fluctuations is that the actual value of exports fell 20 per cent
below the value predicted by the model of secular time trend (equation 1.1)
in eleven of the fifty years covered.30 As the structuralists claim, it would be
difficult correctly to anticipate the availability of foreign exchange if exports
differed greatly from their anticipated levels. However, such fluctuations in
exports did not increase significantly in the Phase V years 1860—77, contradicting the structuralists' contention.3'
ii. The results for imports are similar. The mean absolute proportional
deviations of the level of actual import values from the levels predicted by
equation 1.2 are 0.09 (a=0.08) for 1844-60,0.10 (0.10) for 1861—78, and
0.20 (0.20) for 1878—1927. Once again, fluctuations did not increase significantly between 1844—60 and 1861—77 (test statistic = 0.3), but they did
increase significantly between 1844—77 and 1878—1927 (test statistic = 2.7).
S
AN OVERVIEW
r
e-
12
p
L.
17
iii. An analogous calculation for the mean absolute proportional deviation from a secular trend for ordinary government income, however, gives 0.15
= 0.15), and this figure is not significantly different from
for 1878—1927
the figure of 0.10 (o= 0.09) for the earlier period (test statistic = 1.6). The
structuralists' contention is, therefore, not supported: the significantly greater
fluctuations in exports and imports after the War of the Pacific did not cause
significantly increased fluctuations in government revenue, which would have
had inflationary effects because the government would have had to finance its
debts locally, since it could not readily reduce its expenditures.
iv. After the War of the Pacific, inflation apparently was higher than before, although the evidence is quite fragmentary, but lower, nevertheless, than
the rates observed after the Great Depression (Table A.!, line 2.1). Since
substantial portions of these relatively moderate rates were due to such factors
as wildcat banking, supply disruptions, and internal deficit financing originating
in the Civil War of 1891, the structuralist claim that substantial price instability
originated in the foreign sector seems to be overstated.
v. Of course, price changes are but one aspect of cyclical fluctuations.
Perhaps of more importance are fluctuations in the utilization of productive
capacity. The mean rate of capacity utilization for the 1908—27 period was
93 per cent, an average level not exceeded again until the 1960s for the periods
included in Table A.1 (line 2.2). Thus, on the average, at least during the
twenty years before the Great Depression, no evidence exists of unusually
large forgone output due to fluctuations in foreign markets (or to any other
possible source). However, international market conditions were associated
Id
—ly
•
with the two lowest troughs in capacity utilization in the period: in 1914—15,
because of wartime shortages of imported intermediate supplies, and in 19 19—
22, because of the 1921 depression.32
1.2.5
Liberalization and Economic Growth.
Chilean analysts have presented two quite different characterizations of
growth before the Great Depression. Conflicting policy recommendations have
been made because of the different interpretations of that experience. On the
one hand, Hurtado [1966:57], Lagos [1966:38—39], Pinto [1962:16], and a
number of structuralists contend that the relatively important role of the gov-
ernment in general and of government controls on international economic
transactions in particular before 1860 resulted in much greater industrialization and economic development than occurred in the more liberalized period
after 1860. In contrast, de Castro and de la Cuadra [1971: 1] describe the
decades immediately prior to 1929 as having one of the highest growth rates on
record "in an environment of free trade which made it possible to take advantage of the opportunities created by the international markets while
I
18
INTRODUCTION
demanding, at the same time, efficiency in the domestic allocation of productive
resources."
The conclusion that relatively high growth rates were experienced before
1860 as a result of the restrictive international trade regime seems doubtful
from a number of points of view.
i. The characterization of rapid over-all economic growth is based primarily on the growth of exports, imports, and government revenues. But the
presumption that these growth rates were unusually high before 1860 is not
well founded. For exports the pre-1860 growth rate was found to be significantly higher than the post-1860 growth rate only at the 15 per cent level in
relation 1.1. Again, for government revenues, the difference in growth rates
between the two periods was not significant even at the 25 per cent level.
Moreover, the rates of growth of exports and imports for 1844—60 that are
implied by relations 1.1 and 1.2 were exceeded in a number of periods after
the Great Depression (Table A.l, lines 1.2.4.1 and 1.2.4.2).
ii. Subsequent experience (including that during 1908—27, when imports
and exports were much more important relative to over-all product than in the
more agrarian and self-sufficient economy before 1860) indicates that the relations between the over-all rate of growth and the rates of, growth of exports
and of imports are not simple ones characterized by high correlations.33
iii. Much of the growth which did occur apparently originated not in industry, but in agriculture and mining, partly because of export opportunities
in these sectors.
iv. The industrial expansion may have been an adjustment from a below-
"normal" industrial sectoral share (in the Chenery [1960] sense) resulting
from the effects of the Spanish colonial mercantilist policy to a "normal" share
which would have occurred under a wide range of international economic
regimes once Spanish control was eliminated. If so, subsequent slowing down
of industrial expansion (if indeed it did occur34) may only have indicated that
the relatively easy "catching up" period was over.
v. The existing data refer primarily to the years 1844—60, when substantial liberalization of international trade was occurring. Perhaps, therefore, the
rapid rates of growth of imports and exports only reflected increasing propensities to trade because of the trade liberalization or high income elasticities for
foreign trade.
On the other hand, even if such propensities were not increasing and income elasticities were not high, with the result that equivalent rates of growth
of national income did occur, to attribute that growth to the liberalizing of
foreign trade policy seems at least as consistent with the available data as to
attribute it to the earlier, somewhat protectionist policies.
For the two or three decades preceding the Great Depression, more evidence is available about the probable relation of the international economic
'I,.
AN OVERVIEW
19
regime to growth. For savings, the impact was mixed, in that the effect of the
regime was a continued shift of resources to the government, which apparently
however, in addition, under that regime
had a fairly high propensity to
a level of negative net foreign savings was tolerated which on average was not
The effect
exceeded in any period before 1965 (see Table A.1, line
of the regime on investment seems to have been positive, generally resulting in
a shifting of resources from sectors of average labor productivity37 to higher
ones (with the one exception noted below) Relative domestic efficiency was
raised by subjecting local industry to foreign competition (at least according to
de Castro and de la Cuadra [1971: lJ). In addition, in the early twentieth century, new technology and substantial amounts of foreign capital were allowed to
enter, so that copper mining could be developed on a large scale.
On the other hand, the regime seemed to have negative consequences be-
•
cause it induced the movement of some resources into the manufacturing
tie
hs
pg
•
c
•
it
sector, in which productivity was relatively low.39 Moreover, the development
of the largely foreign-owned export mining sector was allowed to proceed with
little regulation to protect the interests of Chile in cases where these interests
might differ from those of the owners. Finally, excessive upper-class consumption and resource transfers abroad were allowed. As a result, the returns
from the mineral export booms, which might otherwise have been utilized for
substantial capital accumulation, were dissipated.4° The net result of these internal and external factors was a mean annual growth rate in real per capita
product of 1.5 per cent for 1908—27 41 and, probably, a rate of about the same
magnitude for the preceding decades.42 Such a growth rate is not low for the
Chilean experience, but it is exceeded by the rates recorded in several subsequent periods (Table A.1, line 5.1). Even if one is willing to assume a very
strong relationship between trade policies and growth, therefore, this order of
magnitude seems to justify neither the Hurtado-Lagos-Pinto characterization
of that relationship as low and due to the failure of free trade, nor the de
Castro-de la Cuadra claim of it as high and due to the very positive effects of
free trade in the years before the Great Depression.43
C
1.3 THE IMPACT OF THE
GREAT DEPRESSION
The Great Depression began to affect Chile in 1930. The resulting substantial
change in the over-all economic situation is clear from the following observations. In 1930 the capacity to import fell 28 per cent,44 the real value of im-
ports fell 13 per cent, capacity utilization fell 11 per cent, and real mining
product fell 27 per cent (with declines of 32 per cent in copper production,
47 per cent in nitrate production, and 25 per cent in nitrate employment);
20
INTRODUCTION
per capita real GDP fell 14 per cent. Sectoral increases in real product per
capita occurred only in agriculture and in government. Because of "misguided
stubbornness in defending the gold standard" (Hirschman [1963:179]),
monetary policy was contractionary because of a 29 per cent decline in gold reserves. Credit was restricted, wholesale prices declined 13.4 per cent, the real
Central Bank interest rate increased 3.4 per cent, and the NER remained
constant. The only significant anticyclical policy was an increase of 10 per
cent in real government production, and this was financed by foreign
borrowing.45
The extensive negative impact of the Great Depression on the Chilean
economy in 1930 was intensified even further in 1931 and 1932. By the latter
year, in comparison with 1929, the export capacity to import was 18 per cent,
the real value of imports was 13 per cent, nitrate production was 23 per cent,
nitrate employment was 15 per cent, copper production was 32 per cent,
capacity utilization was 55 per cent, and per capita GDP was 49 per cent.48
Even granted that 1929 was a boom year for the Chilean economy, and even
though authorities differ in their estimates for this period (see, for example,
Mamalakis [1971b:40; 1965:384] and Ballesteros and Davis [1963]), the
conclusion that the declines were catastrophic is incontestable on any basis.
Indeed; a League of Nations report characterized Chile as the hardest hit of
any country by the Great Depression.
For the purposes of this study the most important reaction to the Great
Depression was the abrupt change in the prevailing economic philosophy and
policy from liberalism to restrictionism and, somewhat more slowly, to interventionism. Tremendous frustration resulted from too much dependencia on
the external sector.47 In hopes of reducing that dependence and the associate'i
vulnerability, acceleration of internal industrial development was desired.
Restrictive policies, especially for the foreign sector, were seen as the most
effective means of encouraging domestic industry, reducing dependence on the
foreign sector, and conserving scarce foreign exchange. Therefore quite restrictive policies were imposed in 1931.
Before the Great Depression, as is noted in the previous section, some
tendencies toward greater control of international economic relations and reduced dependence on the foreign sector had existed. For example, industrial
import substitution apparently was a major aim of the 1897 and 1928 tariff
laws and the 1925 tax legislation served to lessen the dependence of government revenues on the foreign sector. Despite these elements of continuity,
however, the periods before and after the Great Depression differed substantially in the range and nature of the policy tools used. For the eighty years
before the Great Depression, the major policy tools were a unified NER and
indirect taxes on foreign trade; the characteristics of phases IV and V dominated. For the forty years after the Depression, the economy was characterized
•
•
•
•
flT
prohibited lists, prior deposits on imports, special regimes for exports and
ii
imports, explicit and implicit subsidies, tax rebates, bilateral and compensation
agreements, and regulations concerning direct investment and related flows
and capital movements. Not only was there great expansion in the variety of
policy tools utilized in the attempt to run the more highly controlled system of
d
•
$n
the post-depression era, but also there were great increases in both the fre-
in
quency of policy changes and specificity of policy. The Great Depression, thus,
delineated the most important break in the history of the Chilean experience of
quantitative restrictions.
•
1.4
1.4.1
PHASES OF EXCHANGE CONTROL
AFTER THE GREAT DEPRESSION48
1931—55: Phases I and II—A Quarter Century of
Quantitative Restrictions.
Phase I restrictions were imposed in 1931 and followed by twenty-five
years of Phase II ad hoc adjustment in the foreign trade regime in attempts to
offset the perceived negative results of the restrictions. The degree of quantitative restrictions was relatively constant throughout the quarter century, but
Four subpolicy in other areas of economic activity was less
of the basically Phase-Il-type foreign-sector regime may be dis-
at
r-
•
21
by the more restrictive phases—I, II, and 111—except for a brief experience of
Phase IV in 1959—61. The major policy tools (see Part II, below) included
multiple exchange rates, a myriad of indirect taxes and surcharges on imports,
direct taxes on the major export producers, licensing, quotas, permitted and
d
•
AN OVERVIEW
-5t
tinguished: 1931—39, 1940—46, 1947—51, and 1952—55.
1931 TO 1939.
U
if
•
The first years were characterized by the sharp economic decline discussed in the previous section. The subsequent recovery was only partial: by
1939, real per capita GDP was 76 per cent of the 1930 level and 66 per cent
of the peak-1929 level, and the value of exports in constant United States
dollars was 53 per cent of the 1930 level and 29 per cent of the peak 1920
The resulting subphase means for the growth rate in real per capita
GDP (Chart 1.1) and for the level of capacity utilization were the lowest
among all post-1907 phases and subphases. The subphase mean for the index
-
of export capacity to import was tied for second lowest (Chart 1.1). Moreover, even though prices declined initially (as mentioned above), the rate of
inflation rose to a mean level of 8.5 per cent (Chart 1.1) •52 Despite definite
improvements since the early l930s, therefore, at the end of this subphase the
S
I
I
I
22
U-
INTRODUCTION
over-all economic situation remained substantially worse than that which had
prevailed before the Great Depression.
1940 TO 1946.
This subphase was dominated by two features: First, the Popular Front
government pursued a generally expansionary policy, especially in regard to
the activities of CORFO (Corporación de Fomento), a quasi-public development corporation established in 1939. Second, World War II had the effect of
substantially increasing the demand for Chilean copper and reducing the
availability of imports.53
Economic activity increased considerably because of heightened aggregate demand. Mean capacity utilization returned to its pre-Depression level,
which was not exceeded again until the 1960s (Chart 1.1). The mean rate of
growth of real GDP per capita was the highest recorded in 1908—64 (Chart
1.1). The mean annual rate of growth of real industrial product, 7.5 per cent,
was the highest attained in 1908—70 (line 5.2 in Table A.1).
The factors leading to these gains in real terms, however, also led to
substantial inflationary pressures. Aggregate demand pressures, moreover,
were reinforced by worldwide price increases: over the seven years 1940—46,
unit import values in foreign currency and the imported component of the
wholesale price index increased at mean annual rates of 15.2 and 38.8 per
cent, respectively. The mean annual rate of change in the GDP deflator
increased to 15.1 per cent (Chart 1.1).
Despite the much higher physical quantities of copper exported, the
index of export capacity to import fell slightly to the lowest mean value among
the periods for which data is available, 1928—66 (Chart 1.1) This decline
reflected the combination of fixed copper prices in 1942—46 (due to an agreement with the United States as part of the war effort) and rapidly rising import prices (line 1.1.5.1 in Table A. 1). The PLD-NER, moreover, dropped
almost 40 per cent because devaluations did not match the internal inflation
(Chart 1.1). The exchange control system nevertheless loosened up somewhat. One possible indication is that the ratio of the free- or black-market
exchange rate to the average exchange rate actually declined by 11 per cent
(Chart
The lack of more pressure on foreign exchange despite the
declines in the capacity to import and in the PLD-NER was due to two factors
mentioned above—the increase in the international price of imports and the
limited availability of imports—and to considerable CORFO-related net
capital inflows.
1947To1951.
This immediate-postwar subphase was characterized by reduced capacity
utilization and growth, but accelerated inflation at an average annual level of
L.
k
23
AN OVERVIEW
CHART 1.1
Selected Macroeconomic Variables During Phases and Subphases of the
Chilean Exchange Control Regime, 1908—71
Ill- IV
Phases and subphases in Chilean experience
IV
I-Il
II
II
II
Ill
IV
II
III
II
Per cent or index
Jt
'
El
it
B
1908
27
'2830
'3139
'4046
(continued)
'4751
'52- '56-'59- '62- '65- '71
55 586164 70
24
INTRODUCTION
CHART 1.1 (concluded)
Ill—V
Phases and subphases in Chilean experience
I\/
I—Il
II
IV
Ill
II
II
III
II
II
Per cent or index number
I
I
I
I
I
I
I
I
I
120—
-
110—
-
100—
90
..
—
.......
80—
•..'
,utilization for GDP
-
-
.......
70-
Index of export capacity
60
to import
—
-
I'
50—
'I'
\
Il/I
Rate of change in GOP
20 —
'-4- 4-——
10-
-
——
-
0
—10
I
I
190827
'2830
'3139
I
I
'4046
'4751
55 58 61 64
SouRcE: Table A.1, lines 1.1.1, 1.1.3, 5.1, 1.2.6.3, 2.1, and 2.2.
a. Expressed as a multiple of the national accounts rate.
b. January 1, 1959 = 1.00 (see note to line 1.1.1 in Table A.!).
c. 1965 =
100.
I
In 1965—70 phase, average is for 1965—66 only.
I
I
'52- '56- '59- '62-
'65- '7
70
AN OVERVIEW
25
20.8 per cent (Chart 1.1). The stagnation partially reflected the depression in
the world copper market in 1948—49. Some of the inflationary pressure likewise originated in the foreign sector (i.e., the unit value of imports increased
23.6 per cent in 1947 and 16.8 per cent in 1951; the unit value of exports
increased 31.4 per cent in 1947 and 26.5 per cent in 1951). A midperiod
attempt under Finance Minister Jorge Alessandri to revitalize the economy
and reduce inflation was abortive.
constant-dollar value of both exports and imports increased at very
The former rose
initially because of the termination of the fixed copper price, dropped in the
intermediate years because of recessionary conditions abroad, and increased
in 1951 because of the Korean War boom. The latter grew because of a 48
per cent fall in the PLD-NER (Chart 1.1) and pent-up demand from the
war. In fact, imports would have increased even more had not exchange control increased somewhat due to the scarcity of foreign exchange (Table A.1,
lines 1.1.3 and 1.2.6.2.).
high mean rates (Table A.l, lines 1.2.4.1 and
1952 TO 1955.
The predominant features of these years included rising economic and
political instability and rampant inflation. The mean growth rate in per capita
GDP sank to a postwar trough, mean capacity utilization remained relatively
low, and the mean annual inflation rate skyrocketed to 47.8 per cent (Chart
1.1).
Near the end of this subphase, the internal situation deteriorated greatly.
In 1954 and 1955, real per capita GDP decreased by a total of 4 per cent,
capacity utilization fell to an average level of 89 per cent (the lowest level
since the mid-1930s), real wages fell by a total of 17 per cent below the
historic high of 1953, annual strikes and labor conflicts increased an average
of 25 per cent above the previous all-time high of 1953, and the mean annual
increase in the GDP deflator was 71.7 per cent. In May 1954, a major bluecollar strike took place, and a state of siege was declared in September. In
1955, three different ministers of finance held office in the first half of the year.
In July, a general strike occurred, the long-silent Central Bank issued a report
on the inflation which recommended forceful anti-inflationary measures, and
rumors of a military-backed suspension of the Constitution abounded.58
In the external sector, in contrast, the immediate situation was relatively
favorable in many respects. The mean export capacity to import increased 29
per cent (Chart 1.1). The mean deficit on the goods and services account was
the smallest in absolute value in the postwar period (Chart 1.2). In 1954 and
1955, furthermore, the constant-dollar value of net foreign loans to the
Chilean government increased rapidly. In 1955 exports and imports in con-
26
INTRODUCTION
CHART 1.2
Major Components of Chile's Balance of Payments During Phases and Subphases
of Her Exchange Control Regime, 1947—71
Phases and subphases in Chilean experience
Ill
IV
II
II
II
II
Million 1958 U.S. dollars
I
SouRcE: Table 8.1. Values for 1965—69 and 1970 from that table have been combined in order to obtain the 1965—70 averages shown here.
a. Value for 1971 calculated from data in Table 8.1 under assumption that errors
and omissions were nil.
£
AN OVERVIEW
27
stant dollars rose 18.4 and 11.0 per cent, respectively, and the terms of trade
were the most favorable since 1937.
But the PLD-NER declined to the lowest level of the whole period from
1908 to 1970 because devaluation lagged behind inflation (Chart 1.1). The
resulting high demand for foreign exchange was augmented by the effects of
capital flight prompted by internal instability. Exchange control therefore was
intensified, reaching a postwar high (line 1.2.5.2 in Table A.1). The average
ratio of the black-market to the legal exchange rate reached a record high of
2.66, a level not attained again until 1971 (Chart 1.1).
1.4.2
1956—58: Phase 111—The lbaflez-Klein-Sak.s Stabilization and
Liberalization Program.
Because of the rapidly deteriorating internal situation, in mid-1955 the
government hired the Washington-based Klein-Saks consulting firm to advise
in the formulation of a stabilization program. The effort of stabilization plus
liberalization that commenced in early 1956 was strongly identified with the
Klein-Saks mission.
A number of indices suggest that this stabilization attempt met with at
least initial success. The mean rate of rise of prices was reduced from 47.8 per
cent in 1952—55 to 37.6 per cent in 1956—58 (Chart 1.1), with annual rises
declining from 58.3 per cent in 1956 to 28.3 per cent in 1957 and to 26.3 per
cent in 1958. Although the mean rate of capacity utilization declined slightly
below the 1952—55 average, it was higher than the 1954 and 1955 levels
(Chart 1.1). The mean rate of growth of per capita real GDP was the highest
recorded among the phases and subphases since the end of the Second World
War (Chart 1.1), as a decline in industry was more than offset by increases in
agriculture and other sectors (lines 5.1, 5.2, and 5.3 in Table A.!). However,
in most respects, the economy worsened in 1958.
In the foreign sector, liberalization resulted in a shift to a Phase III
regime. Accompanying this liberalization was a fall of 46 per cent in the mean
ratio of the black-market NER to the over-all average NER, a rise of 30 per
cent in the PLD-NER, and an increase of 5 per cent in the mean export Ca-
pacity to import (Chart 1.1). Imports grew so much more rapidly than exports, however, that the mean deficit on the goods and services account in
millions of 1958 dollars increased from —5.6 to —81.7 (Chart 1.2). This
enlarged deficit was partially offset by mean increases in net unrequited transfers from —0.1 to 19.6 million 1958 dollars (largely due to foreign official
support for the stabilization effort) and in net nonmonetary capital inflows
from 18.6 to 37.4 million 1958 dollars (Chart 1.2). But mean net monetary
capital and gold movements also increased—from —5.6 to 32.7 million 1958
dollars (Chart 1.2)—which placed growing pressure on the dwindling foreign
28
INTRODUCTION
exchange reserves. Among the causes of this pressure were a reduction in export value due to a decline in the terms of trade, which reflected recessionary
conditions in foreign markets (line 1.1.5.1 in Table A. 1) and increased demand
for foreign exchange for imports due to the decline in the PLD-NER after the
fixing of NERs at constant levels in mid-1957. As a result of the increasing
pressure on foreign exchange reserves, exchange controls were intensified in
the last half of this phase, although probably not to the same degree as in 1955.
1.4.3
1959—61: Phase IV—The Alessandri Stabilization and
Liberalization Policies.
In the presidential election of late 1958, Jorge Alessandri, a conservative,
.narrowly defeated Senator Salvador Allende Gossens, the Socialist candidate.
The new government had no formal stabilization program, but generally hoped
to achieve price stabilization and to revitalize the economy through increased
supplies of all goods, to be induced by liberalized internal and external policies.
In comparison with the Klein-Saks program of the previous phase, the Alessandri government seemed to place revitalization above price stabilization °
and to place greater importance on the role of a liberalized foreign sector.
The Alessandn policies had mixed results during this period. The mean
capacity utilization rate was the highest since the Second World War, although
lower than rates attained in the 1960s and early 1970s (Chart 1.1). The mean
rate of inflation was the lowest in the entire postwar period, 1946—71 (Chart
1.1). In all the twenty-six postwar years through 1971, rates as low as the ones
in 1960 and 1961—9.5 and 7.9 per cent—were recorded only two other
times. The rate of growth of real GDP, however, fell to the lowest mean level
of any phase after 1955, although it did exceed the means for the subphases
between 1928 and 1940 and 1947—55.
Liberalization of the foreign sector proceeded steadily. The result was a
Phase IV regime, the most liberal since the Great Depression. One index of the
relative lack of restrictions is the drop in the mean ratio of the black-market
NER to the average legal NER; it fell 30 per cent to a post-Depression low of
approximately 1.0 (Chart 1.1 and line 1.2.6.2 in Table A.1).
The mean export capacity to import rose by 9 per cent (Chart 1.1) despite a small decline in the mean terms of trade (line 1.1.5.1 in Table A.1).
The fall in the mean PLD-NER, which resulted from the maintenance of a
fixed NER after the January 1959 devaluation (Chart 1.1), and the reduction
of restrictions on imports, however, resulted in an increase of over 18 per cent
in the mean value of imports in constant dollars (line 1.2 in Table 8.1). The
mean deficit on the goods and services account, therefore, rose from —81.7 to
—163.0 millions of 1958 dollars (Chart 1.2). Despite partially offsetting increases in mean net unrequited transfers, which resulted from international
L
AN OVERVIEW
29
support of the stabilization program, and in mean net nonmonetary capital
inflows, which resulted from new foreign investment and capital repatriation,
compensatory mean net monetary movements reached the highest level of the
1947—70 period (Chart 1.2). Mean net foreign reserves held by the banking
•
system fell from $29.9 million in the previous phase to —$60.0 million in this
one (line 1.2.6.4 in Table A.1). At the end of 1961 these reserves amounted
•
5.
to —$164.8 million.
1.4.4
1962—64: Phase Il—Reversal to Restrictionism, Interventionism, and
Inflation.
In December 1961 the Central Bank's liquid foreign reserves were equal
to the value of about ten days of imports, a considerable number of import
applications were pending, and the Central Bank could no longer borrow
abroad because of the substantial outstanding foreign debt. Near the end of
that month, therefore, the Central Bank revoked all grants of authority to deal
in foreign exchange and established temporary 10,000 per cent prior-deposit
requirements on all imports. The Alessandn stabilization attempt and the Phase
IV type of foreign-sector regime thus came to an end. The next three years
11
is
a
e
a
•
•
were characterized even less than the 1959—6 1 period by a systematic over-all
economic program. Instead, piecemeal policies led to a generally more complicated and restrictive system.
The general performance of the economy in 1962—64 was in some respects quite satisfactory. The mean rate of growth of real GDP was the highest
since the Second World War (Chart 1.1). The mean rate of capacity utilization
was the highest it had been in any phase or subphase from 1908 to 1964 (Chart
1.1). However, there were problems. The mean annual rate of change of prices
increased to 35.4 per cent (Chart 1.1), with levels recorded in 1963 and 1964
that were exceeded in the Chilean experience only in the near-hyperinflations
of 1954—56 and 1972—76. The relatively high mean rate of growth, moreover,
was due almost entirely to events in 1962. On a per capita basis, the economy
in real terms almost stagnated in 1963 and 1964.
Inflation outpaced even the rather substantial devaluations; so the PLDNER fell another 5 per cent (Chart 1.1). The increased inducement to import
was restrained only by the return of quantitative restrictions to a Phase II level.
The mean ratio of the black-market NER to the average of the legal NERs
rose to 1.70 (Chart 1.1), providing one measure of the increase in exchange
control. Bacha and Taylor [1963: 144] estimate that the degree of overvaluation averaged 47 per cent in this phase.
The export capacity to import increased about 20 per cent (Chart 1.1)
because of an improvement of almost 10 per cent in the terms of trade (line
1.1.5.1 in Table A.!) and an expansion in the physical quantity of copper
30
INTRODUCTION
exports. Even though exports grew more rapidly than imports, however, the
mean net deficit on the goods and services account rose to its largest value in
the postwar period primarily because of the large augmentation in net factor
payments abroad (Chart 1.2, and line 5 in Table 8.1). Mean net unrequited
transfers fell, but mean net nonmonetary capital inflows increased substantially
because mean net inflow of loans to the central government was the highest
recorded in any period since World War II (Chart 1.2 and line 13.1 in Table
8.1). As one result, mean net monetary capital and gold movements fell to
almost half the level of the previous period (Chart 1.3). As another result,
foreign debt by the end of 1964 had risen to $1,896 million—278 per cent of
the export value or 36.1 per cent of GDP for that year.
1.4.5
1965—70: Phase Ill—The Frei Stablization and
Liberalization Program.
In the presidential election of late 1964 the Christian Democratic candidate, Eduardo Frei Montalva, won by a substantial margin (the runner-up,
once again, was Senator Salvador Allende, the leftist-front candidate). The
new government had as its goals substantial changes in the economic, political,
and social fabric of the country. Among the major objectives were (i) increases
in the rates of economic growth through higher capacity utilization and Capacity expansion; °° (ii) gradual programmed price stabilization; (iii) redistribution of income and of political power toward the lower economic classes;
(iv) increased national control over the national destiny and the elimination of
balance of payments problems by changing the nature of control over Chilean
mineral resources, by increasing the capacity to export both traditional and
nontraditional products, and by increasing domestic production in "dynamic"
sectors; and (v) structural changes including reforms in agriculture, social security, education, the Constitution, housing, taxation, government, and (later)
industrial organization; and popular promotion (promoción popular) of wider
participation in decision making.
These broad objectives had firm roots in the structuralist analysis of the
Chilean economy. Price stabilization was included as one of the major objectives for the third time in a decade, but was not given priority over all other
aims. Foreign-sector liberalization was initiated in some ways almost immediately. A sliding-peg NER policy was introduced in April 1965, with weekly
or biweekly devaluations (section 3.2). Exchange control was reduced much
more, however, in the last half of the period, after the Vietnam War—related
copper boom had provided increased foreign exchange and the momentum for
internal reform had died. The result was a Phase III regime over the period
as a whole, with perhaps some Phase IV characteristics appearing in the last
years.
t
t
I
31
AN OVERVIEW
The indication of a number of indices is that the average general economic
performance was relatively satisfactory during the Frei regime (Chart 1.1).
The mean growth rate of per capita real product was the highest attained in
any phase or subphase since 1908. The mean capacity utilization level had
been exceeded only once before. The mean rate of change in the GDP deflator
•
declined significantly from the level of the 1962—64 phase.
Once again, however, the mean values disguise a deteriorating intraphase
situation. Almost all of the per capita real growth, the highest capacity utilizatiOn rates, and the lowest rates of inflation occurred in 1965 and 1966. Thereafter, the economy was relatively stagnant with increasing inflation.
•
The mean value of exports in constant dollars rose almost 60 per cent
because of the copper-market boom of the late 1960s (line 1.1 in Table 8.1).
For the first time, the export capacity to import exceeded the pre-Depression
level (Chart 1.1). Because of the continuing fall in the mean PLD-NER (Chart
1.1) and the reduction in exchange control, mean PLD-EERs dropped considerably. Mean imports in constant dollars increased 21 per cent (line 1.2 in
Table A. 1). The net impact of the changes in exports and imports was a reduction of almost 60 per cent in the mean deficit on the goods and services
account (Chart 1.2), despite an increase of 75 per cent in the mean net outflow
of investment income (Table 8.1, line 5). The reduction in this deficit, together with slight increases in mean net unrequited transfers and nonmonetary
capital inflows, resulted in a drop in mean compensatory capital and gold
movements to post-World War II lows (Chart 1.2). By the end of this phase,
net international reserves of the banking system had reached the highest level
in at least two decades. But the foreign debt nevertheless grew at an average
rate of 7.3 per cent per year to a level of approximately $3 billion by late
1970—248 per cent of annual export value or 27.2 per cent of GDP.61 The
debt position thus was ominous despite the benefits from the copper-market
L.
•
•
-
boom.
)
e
r
•
i
1.4.6
1971—73: Phase lI—Allende and Popular Unity Policies.
In the presidential election of late 1970 Salvador Allende, the candidate
of Popular Unity (a coalition of the Communist, Socialist, Radical, and several
minor parties), narrowly defeated former president Alessandri (thus reversing
the results of the 1958 election). According to Américo Zorrilla [1970:1477],
the first minister of finance, the central long-run aim of the new government
was "to transform the present economic structure, terminating the power of
foreign and domestic monopolistic capital and of the latifundia, in order to
initiate the construction of socialism." 62 Related medium- and long-run goals
included: (i) the redirection of benefits of the economy more to lower- and
middle-income classes; (ii) the development of an extensive multilevel plan-
-
.
-
P
32
INTRODUCTION
ning system with wide participation; (iii) the reduction of external dependence
in all of its possible forms; and (iv) the establishment of a new ownership structure for the means of production—state (basic natural resources, large monopolies, banks, foreign commerce, and other strategic activities for development),
mixed (parts of the industrial, fishing, mining, and trade sectors), and private.
The government proposed to move toward these goals gradually and constitutionally in "the Chilean way," not by extralegal means.
To broaden support for the long-run attempt to transform society, the
government recognized that some short-run successes were essential. The cornponents of short-run policy, as outlined in the 1971 plan, were the following:
i. Revitalization of the economy by reducing uncertainty through the use
of guaranteed production contracts with the state and by increasing aggregate
demand through increased exports, a substantial redistribution of income, and
a large increase in government expenditures on public works, housing, and
human capital;
ii. Constraint of inflation by strict price controls, by freezing the NER,
and by increasing supplies from domestic production (by means of increased
capacity utilization) and from imports (which would be financed through expanded exports and the available international reserves accumulated during
the 1960s copper boom);
iii. Redistribution of income to the lower and middle classes by a minimum readjustment of wages and salaries of 100 per cent of past price increases
(and larger adjustments for lower wages and salaries) in combination with
reduced inflation; by increased lower-income housing; by increased humancapital investment (especially in education and in health); by increased employment; by reductions in returns to national and foreign higher-income groups
through the nationalization of key subsectors in the economy; by accelerated
agrarian reform; and by the democratization of the distribution of credit;
iv. Redistribution of control over resources from the private to the public
sector in hopes of providing the basis for planned "harmonized and balanced
growth" which would favor the lower and middle classes by expanding the
planning apparatus and by extending state control into the productive sector;
v. Initiation of structural reform, which would lay the basis for high longrun growth and for the transformation to socialism by means of accelerated
investment of public-sector
by the enlargement of the role of the
state in the economy, and by acceleration of agrarian reform;
vi. Expansion of state activities in all aspects of international economic
interchange by the reintroduction of a Phase II regime in order to reduce foreign dependence of all types and to utilize the foreign sector in the pursuit of
the other aims mentioned above.
In 1971 the performance of the economy was quite positive in a number
of important general respects. Capacity utilization rose to a very high level
St
of
th
al
W
ar
m
01
f
d
5
f
1'
in
m
tic
th'
r
(1
lei
th
na
no
AN OVERVIEW
33
(Chart 1.1). Unemployment dropped substantially (Table A. 1, line 2.3).64
The rate of change of prices fell to 20 per cent (Chart 1.1). Imports in constant dollars increased over 6 per cent (Table 8.1, line 1.2). The rate of growth
of per capita real product soared above the means of any phases or subphases
since 1907 (Chart 1.1) and exceeded annual means of every year but one after
be
se
the Second World War.
In the same year the new government also succeeded in several ways in its
aims of transforming ownership relations and of extending state control. With
unanimous legislative support, Chile completely nationalized large-scale copper
mining, which previously had been subject to considerable foreign control.°5
The state extended its direct control over a number of other formerly foreignand domestically-owned enterprises, including most private banks. The government accelerated agrarian reform. Exchange control increased, and the publicsector share in approved import registration rose from 34 per cent in 1970 to
44 per cent in 1971 (and to 65 per cent in the first eight months of 1972).
Some of these short-term gains, however, were purchased at significant
long-term costs:
i. The high growth in 1971 reflected higher capacity utilization resulting
from expansion of current consumption more than it did an increase in pro-
ductive capacity. In fact, in that year the rate of augmentation of capacity
declined. Real physical investment dropped by 7.7 per cent (Table A.!, line
5.4),66 and the constant-dollar value of imported machinery and equipment
th
Ii-
fell 16.8 per cent°7 The fall in the rate of growth of industrial production from
values ranging from 10.8 to 14.7 per cent in 1971 to 2.5—2.8 per cent in 1972
may reflect in part the approach of the economy to capacity limits.88
ii. The decline in the rate of change of prices in 1971 occurred despite a
121 per cent increase in the money supply because of greater supplies of goods
from domestic production and imports, more extensive price controls, the
maintenance of a fixed NER, and an upward shift in the demand for real
monetary balances resulting from income redistribution and changed expectations. Given the augmentation of the money supply and of labor payments,
these measures could not suppress inflation for long. In late 1971, the pace of
price increases began to accelerate. In 1972 the consumer price index rose 178
per cent, an all-time high. Inflation increased even more in 1973. Black markets flourished for many commodities.
iii. The constant-dollar value of exports dropped 12 per cent in 1971
•
(Table 8.1, line 1.1) and declined again in 1972 because of the slump in the
70 and the failure of Chilean production to reach planned
world copper
levels (e.g., in 1971, production was less than 80 per cent of that planned for
the year). In part because of United States reaction to the terms of the copper
nationalization, net unrequited transfers in constant dollars also fell, and net
nonmonetary movements in constant dollars plummeted (Chart 1.2). At the
34
INTRODUCTION
same time imports in constant dollars increased over 6 per cent in 1971 (Table
8.1, line 1.2) and approved import registrations apparently rose even faster in
1972.71 As a result, international reserves held by the banking system fell by
over $300 million in 1971, net monetary capital and gold outflows in constant
dollars reached a post-World War II high (Chart 1.3), and a severe foreign
exchange shortage existed by the end of 1971. Therefore, the escudo was devalued by 30 per cent in December 1971 and by 58 per cent in August 1972,
exchange control became much more restrictive, and the large foreign debt
obligations were renegotiated in 1972.72 Nevertheless, the country continued to
be on the verge of foreign-exchange crises.
iv. The attempt to transform the power and ownership relations of society
caused considerable politicization and polarization in Chilean society. In 1972
martial law had to be invoked several times, and the economy ground to a halt
in October because of a 25-day work stoppage by the Truck Owners' Association (supported by many business and professional groups) which was termi-
nated after the government added three military members to the cabinet. Internal instabilities increased in 1973. Copper miners went on strike in May. An
attempted military coup was squelched on June 29, 1973. Truckers, businessmen, and professionals again struck. On September 11, Allende died in a military coup.
1.5 MAJOR CONCLUSIONS
Not all of the detailed results of this book can be presented here in a sufficiently
brief form. The following points, however, integrate and summarize many of
the major thrusts of this analysis of the Chilean experience.
•
•
•
i. Early origins of exchange control and import substitution. The introduction of exchange control as an element of Chilean international economic
policy was not a post-World War II phenomenon, as apparently was the case
in a number of other developing economies. Instead, exchange control was
introduced in the early 193 Os in response to the devastating impact of the Great
Depression on a previously quite open economy.73 Moreover, fairly continuous
conscious attempts to industrialize by import substitution dated back at least to
the 1897 tariff law (and, perhaps with less continuity, to before 1860). Most
of the aggregate industrial import substitution which had been accomplished by
the mid-1960s had in fact been achieved before the end of World War II. Thus,
although the exhaustion of foreign reserves accumulated during World War II
and the end of the Korean War boom intensified exchange control and efforts
to induce import substitution, both of these phenomena originated much earlier.
ii. Cyclical pattern of restrictions and overvaluation in a system in disequilibrium. After the Second World War disequilibrium combined with an
—4
AN OVERVIEW
le
y
it
to
2
Ii•
of
ic
at
35
overvalued NER prevailed because of the perceived negative impact of devaluation on income distribution, inflation, and the cost of imported capital goods.
The intensity of exchange control, however, varied considerably. The system,
in combination with external stimuli, generated cycles in the degree of cornplexity and restrictiveness. Subsequent to devaluation and rationalization, ad
hoc changes of greater specificity were made with increasing frequency because
of the perception of unintended side effects of quantitative restrictions, the
attempt by the government to use such restrictions for a multitude of reasons
additional to managing the balance of payments, and the growing overvaluation due to internal inflation. In each cycle the government resisted formal
devaluation for some time because of the perceived negative impact of the
action. However, inflation was so fast that, despite accelerated ad hoc changes
iii Chile's international economic policy, devaluation of the NER(s) to new
fixed rates (generally accompanied by considerable rationalization and some
liberalization) could not be avoided. At that point the cycle would begin anew.
Moreover, even when the government operated a sliding-peg NER policy with
frequent devaluations, the NER(s) remained overvalued so that the cycle was
only moderated, but not eliminated.
iii. Motivations for shifts in exchange control. External developments
have been influential in the decisions to alter foreign-sector policies. Extensive
exchange control was reintroduced in reaction to the Great Depression. No
other course was judged to be possible at the time. The experience of the 1930s
also led to some general attitudes which have shaped the evaluation of policy
possibilities in the ensuing decades: export pessimism and a desire to establish
greater domestic control over the national economic destiny, even at the cost
of forgoing possibly considerable static comparative advantage. Specific policies, however, often have been ad hoc responses to immediate foreign market
conditions. Precarious foreign-exchange situations have developed repeatedly
because of the disequilibrium system with its limited inducements for export
expansion and extensive needs for intermediate and capital goods imports for
the import-substitution industries. Instead of establishing greater independence
from the international market, therefore, the trade regime in some respects
caused the economy to become more vulnerable to foreign-market fluctuations.
Thus, restrictiveness tightened during slumps in the world copper market and
loosened during booms in that market (such as those related to the Second
World War, the Korean War, and the Vietnam War).
I
S
•
-
The importance of external conditions, however, should not be overstated.
Most of the recent shifts from one phase to another have been but a part of
much larger policy programs for which the dominant causal factors have been
internal political changes and policy considerations (especially those concerning cyclical fluctuations and income distribution).
iv. Nominal versus effective devaluations. Except when the sliding-peg
36
INTRODUCTION
exchange-rate policy was in effect, in 1965—70, effective devaluations were
smaller in magnitude than the nominal devaluations from which they resulted
because devaluations usually have been accompanied by partial removal of
import restrictions, surcharges, and export subsidies. Another factor of considerable importance, however, is that devaluations have led to substantial and
fairly immediate responses in the domestic price level. Thus, care must be
taken in the assessment of liberalization attempts not to confuse formal alterations in exchange rates with effective devaluations.
v. Problems in evaluating the impact of foreign-sector policies. The difficulties in assessing the effects of foreign trade and payments policies are many,
as is well-known. Several, which are made quite clear in the present study,
however, merit mention because they are often ignored in other analyses. (a)
Partial-equilibrium conclusions about the impact of changes in the magnitude
and, in some cases, even the direction of the foreign-sector regime may be very
misleading. The general-equilibrium analysis below, for example, suggests that
the impact of devaluation on the balance of payments and on employment is
much less positive than has been suggested by partial-equilibrium analyses
because of indirect offsetting effects transmitted in large part through the monetary and price system. (b) The time patterns of responses to policy changes
vary considerably from one area of concern to another. In. some cases, moreover, the direction of the response reverses over time. An evaluation of the
success or failure of such policies, therefore, may depend crucially upon what
time period is relevant. (c) If the empirical measures derived from the Chilean
experience are appropriate indicators, substantial effort may have been mis-
ç
b
t
placed in the considerable recent literature on alternative protection and
resource-cost estimates. Intrameasure differences in the treatment of the cost
of capital, for example, apparently are about as important a source of disparities among sectors as are the much more debated distinctions between
EPRs and DRCs. Some of the other intrameasure concerns which have received relatively great emphasis (e.g., the treatment of nontraded inputs),
moreover, do not make much difference. Within a general-equilibrium context, finally, the pattern of changes in the prices ratios of final to intermediate
products frequently differs from patterns of changes in either capacity utilization or capacity creation. Measures related to the price ratios, therefore, may
be quite misleading indicators of short- and long_run resource shifts.74
vi. Wide variations in economic costs. Variations in DRCs are quite large
both between and within sectors. Some industries appear to be quite efficient,
while others require a large multiple of all resources in order to save an identical amount of foreign exchange. This large variation apparently reflects the
failure of the exchange control system to indicate differences in social profitability to individual decision makers. Industries are encouraged indiscrimi-
o
5
t
s
s
t
f
AN OVERVIEW
e
d
i.
•
.d
•
a-
•
ty
•
'.
es
37
nately. The result is a much higher social cost than might obtain if the low-cost
industries could be given preference.
vii. Short- and medium-term constraints on exchange liberalization. In
light of the short- and medium-term impact (i.e., within one to three years)
on major areas of the economy, the repeated reluctance of Chilean governments to devalue and to reduce exchange control is understandable, especially
if a high enough discount rate is used. In the short run, the results of tightened
quantitative restrictions have been to enlarge the net command over international resources from trade; extend national autonomy by some measure; increase capacity utilization; and favor industry over other goods sectors and
goods over services. In the short run, devaluation makes a smaller positive
contribution to the balance of payments than is often suggested, causes an
immediate reduction in capacity utilization, has a substantial inflationary impact, is discriminatory against industry, and has substantial unwanted effects
by shifting control over income and over resources away from labor and the
government and toward foreigners.75 In many respects specific liberalizations
adopted toward previously foreign-owned large-scale mining, e.g., the Nuevo
Trato of 1955, which is discussed in subsection 4.2.1, did not on balance work
to Chile's advantage.
Exchange control to maintain an overvalued NER is, of course, hardly
without undesirable features even in the short run: Exports have been discouraged.78 Variations in imports have been increased, with disruptive effects
S-
6-
on supplies in wide areas of the economy. The composition of imports has
been altered in unwanted ways. Government deficits have been enlarged be-
cause revenues from exports and imports were reduced;77 as a result, inflation
has been increased and income distribution changed for the worse. Finally,
substantial resources have been needed to operate the system.
Thus the point is not that overvaluation plus exchange control is without
costs, but that there are substantial costs as well as possible benefits to moving
away from the type of payments regime which has been dominant in Chile for
the past four decades. Without a specific welfare function, it is not clear that
the benefits outweigh the costs. The repeated reluctance of Chilean governments to liberalize the foreign sector may be quite rational, or at least understandable. Even if the government is convinced that the long-run net benefits
of eliminating exchange control and overvaluation are positive and substantial,
it might reasonably hesitate to proceed with such changes because of the high
short- and medium-term costs as seen either from the broad viewpoint of the
nation or from the narrower political interests of the governing group.78 Given
the history of past failures of programs of liberalization plus stabilization,
•
future significant liberalization probably will be attempted only if the government is convinced of the following: that the long-run net benefits will be sub-
38
INTRODUCTION
stantial, that it had unusually broad internal political support derived from a
widespread perception that it could resolve a severe economic or political crisis,
and that it had expectations of a very large command over foreign resources
from capital inflows or from a copper-market boom.
viii. Development and exchange liberalization. Many interrelationships
exist between the trade and payments regimes and economic development. Although foreign-sector policies affect many aspects of the economy, the evidence
to be presented in this study does not indicate overwhelming support for either
of two polar positions, (a) that exchange control accelerates development,
since resources can be allocated directly to the uses where developmental payoffs are the highest or (b) that liberalization quickens development because of
international transfer of knowledge, comparative advantage, and market pressures for efficiency. Instead, the evidence is mixed and indicates that both of
these positions may overstate the contribution that changes in the foreign trade
regime could make to development.
It is possible that the underlying evidence is too closely tied to short-run
situations, despite my attempts to incorporate adjustments to longer-run phenomena. A liberalization which was viewed as permanent might have substantially greater positive impacts on growth through large nonmarginal changes
such as Fishlow [1974] reports in the Brazilian case. Perhaps Bruton [1967]
is correct in his speculation that exchange control has resulted in substantially
less productivity growth in Chile and other Latin American countries because
much of the economy has been isolated from international market pressures
for increasing efficiency.
I am fairly sympathetic to the notion that the Chilean exchange control
system may have dampened development achievements. The system has not
seemed sufficiently flexible to adapt to changing conditions. Protection, once
granted, has been very difficult to reverse. In the early 1960s, some of the most
protected industrial subsectors included the traditional ones which had received
deliberate protection—in some cases since the 1897 tariff law—and in which
import substitution had been achieved largely before the Second World War.
The exchange control system seems to have an inner logic of its own which
perpetuates its existence because of the costs of dismantling it despite the limitations it may impose on development achievements.
On the basis of the present study alone, however, it would be primarily
speculative to assert that really substantial development benefits can be attained
through the selection of a particular foreign-exchange regime. Although there
is plenty of evidence of interaction between such regimes and variables which
are widely thought to be related to the development process, the results do not
warrant a firm conclusion about the preferred type of regime. Part of the problem, of course, is that our knowledge of what really underlies the process of
development is limited.
I
39
AN OVERVIEW
a
es
her
of
of
de
ix. Gradual versus abrupt liberalizations. If liberalization is to have a
substantial long-run development payoff, expectations that the policy will be
permanent must be strengthened. Otherwise, efforts will be devoted not to increasing productivity, but to obtaining the private benefits of greater protection.
Given the failure of three liberalizations and four stabilization attempts in less
than two decades, it will not be easy to make new programs credible.
Gradual attempts have the advantage of allowing some short-run gains.
But they probably would not cause sufficient short-run benefits to enable the
maintenance of the necessary political momentum until longer-run returns
became clear. The best hope for a successful liberalization, therefore, probably is an abrupt imposition which incurs certain short-run costs, but also alters
expectations so that longer-run advantages may become obvious before poutical momentum falters.
NOTES
,7]
illy
j.se
L
101
Some analysts have developed the contrary view—that the role of the foreign
sector in economic development is not necessarily positive. In some cases, these views,
which are mentioned below, have been based on the Chilean experience.
2. Different studies have different weaknesses. Those with broader historical and
macroeconomic perspectives often tend to be less rigorous analytically (for example,
Pinto [1962]). Those with a more careful and detailed data analysis often tend to have
a limited historical perspective (for example, Jeanneret [19711). (The bibliography
provided at the end of this book contains publication details for all works cited.)
3. For fuller presentations and critiques of the structuralist analysis, see Baer [1967],
Behrman [1974], Corbo [1971], Edel [1969], Campos [1964], Paws [1972], Prebisch
[1961], Seers [1963], Sierra [1969], Sunkel [1958] and Wachter [1974].
4. Analyses of Chilean economic history which have had substantial impact in
recent years include Ahuinada [1956], Pinto [1962, 1964], Sunkel [1958], and Instituto
de EconomIa [1956, 1963]. Part of this impact is due to the important political and
advisory positions which economists have held in Chile for more than a century. A
ir.
ch
Ii
few examples include Jean Gustave Courcelle-Seneuil (see section 1.2, below); Felipe
Herrera, who served as minister of finance in 1953 in the Ibaflez government; Carlos
Massad and Jorge Cauas, who served as president and vice-president of the Central Bank
in the Frei administration of 1964—70; and Pedro Viiskovic, who was minister of
economics for the first twenty-two months of the Allende government.
5. Basic sources for this section include Ballesteros and Davis [1963), Cohen [1960],
Ellsworth [1945], Garcia [1964], Hirschman [1963], Humud [1969], Hurtado [19661,
Lagos [1966], Mamalakis [1971b], Mamalakis [1965], Marshall [1957], Mufloz [1968],
Pinto [1962], Reynolds [1965], and Rufatt [1972].
6. The phases are defined in Appendix D. These phases are descriptive only; no
causal inference is intended.
7. For a definition of this and all other terms and abbreviations relating to the
foreign trade regimes, see Appendix D.
8. The value of the escudo-equivalent in 1960 U.S. dollars at time t
is
the purchasing
power in terms of 1960 dollars of one escudo (or, equivalently, 1,000 pesos) at time 1.
40
INTRODUCTION
(In 1960, the basic currency unit was changed from the peso to the escudo at a conversion rate of 1,000 pesos per escudo.) For more recent periods, see Table A.1, line 1.L2.
9. Pinto [1962:15], in fact, suggests that the next phase (i.e., Phase V) may have
started as early as 1848.
10. For the decades from 1880 to 1930, Hirschman [1963:160] reports average
annual rates of change of prices ranging from 3 to 8 per cent (see also Table Al,
line 2.1, below). For extensive discussions of the causes of this inflation see Fetter
[19311, Garcia [1964:12—15], Hirschman [1963:160—183], Humud [1969:64—72], Mamalakis [1971b:97—98, 117—120], Marshal] [1957:4—9], and Pinto [1962:59—64].
11. Unless otherwise noted, data in this study on long-run trends in the copper
subsector are based on Senado, Oficina de Informaciones [1971:20—21).
12. The estimate by Hurtado [1966:38—59] is lower than Pinto's. On either count,
however, a substantial postwar decline in the share of agriculture is implied.
13. In fact the foreign-owned share of nitrates increased from 64 per cent in 1882
to 85 per cent in 1901. See Mamalakis [1971b: 105—ItS, 415—498] and Pinto [1962:53—571
for these and other details about nitrates in this era.
14. Mamalakis [1971b:498] estimates that Chilean ownership was 4.5 per cent of
the total in 1918. Reynolds [1965:219] estimates that the Chilean share was 11 per cent
in 1920.
15. For example, before the establishment of an income tax, in 1925, the tax
incidence on the gross value of copper production was 0.8 per cent, and neither copper
nor nitrates were subject to the impact of significant nontax government policies. Even
after the imposition of the income tax on copper, less than 40 per cent of the total
value of production of large-scale copper mining remained in Chile (of which over
three-fourths was for local operating expenses), and less than I per cent of the labor
force was employed in large-scale copper mining. The comparatively low value of series
1.1.5.2 in Table A.1 despite high copper prices also reflects limited linkage. For further
details see Reynolds [1965:223, 226, 376, and 378].
16. For further discussion of these factors, see Gunder-Frank [1969: 17—84, 89—98],
Mamalakis [1971b:105—1I5, 465—498] and Pinto [1962:53—57].
17. Mamalakis [l971b: 106—115] suggests that in the early twentieth century factor
payments abroad for mining (copper as well as nitrates) were largely offset by capital
inflows and probably were less in magnitude than the amounts involved in the flights of
domestic capital resulting from internal instability. This analysis seems to be at least
partially misleading in that the capital inflows to which he refers were primarily for
large-scale copper mining from which substantial net outflows were to occur for at least
four decades subsequent to the end of this period.
18. Mufloz [1968:19] claims that a substantial increase in average tariff rates was
implemented earlier in 1878. Humud's [1969:100, 122. 130] data, however, indicates
that the general legal tariff rate was 25 per cent both before and after 1878 and that
the average actual nominal rate did not increase substantially at that time.
19. These import substitution coefficients are defined as
— u,) (D, + M,)/
— D0), where D equals domestic production, M equals imported supply. ii = MID,
and the subscripts refer to the endpoints of the time period over which the comparison
is made. For a critique of this measure based on the exclusion of impacts on intermediate
products, see Morley and Smith [1970].
20. Mamalakis [1971b:841 suggests that the increased taxes on industrial imports
in the 1830—60 period reflected not a deliberate intent of the government to protect
domestic industry per se, but a desire on the part of the powerful agricultural elite to
shift the tax burden to the economy as a whole. However, considerable evidence exists,
—.--
.-
.
.
.
41
AN OVERVIEW
both for this and for later periods, that protection was an important motivation for
upward adjusments in the tariff structure. For examples, see Ellsworth [1945:49], Jeanneret [1971:145]. and Mamalakis [1971b:141.
21. For the import-to-GDP ratio, Ellsworth [1945:3] presents a range of 0.30 to
0.40 for before the Great Depression, and CEPAL (as reported in Muñoz [1968:831)
estimates 0.31 in 1929. A comparison of the Humud [19691 trade data and the Ballesteros
and Davis [1963] product data suggest that by 1929, the ratio had fallen substantially
from previous levels. In Table Al, for the period from 1908 to 1927, the import ratio
(Line 1.2.6.7) is 0.52, and the export ratio (line 1.2.6.8) is 0.58: both seem somewhat
high. The discrepancy in the same table between the export ratio and the ratio of mining
GDP (line 3.1.1.2) to over-all GDP also raises similar doubts. However, some discrepancies would be expected because of the considerable size of positive net factor
payments abroad and because the GDP estimates are based on real quantity data while
the real value of exports reflects secularly increasing relative prices for nitrates over
the period.
22. Throughout this study all regression estimates presented are based on ordinary
least squares techniques, except when the Cochrane-Orcutt procedure has been utilized
to estimate a first-order autocorrelation coefficient (p). The absolute values of
statistics are presented in parentheses beneath the point estimates.
is the coefficient
of determination (corrected for degrees of freedom), SE is the standard error of estimate, and DW is the Durbin-Watson statistic. The data source for this and all other
regression estimates presented in this subsection is Humud [1969].
23. Additive and multiplicative dummy variables for different phases were excluded
because the coefficients of the dummies were not significantly different from zero even
at the 15 per cent level.
24. A simple autocorrelated time trend model, however, is consistent with only 14
per cent of the variation in extraordinary government income over the sample.
25. Moreover, deliberate decisions were made to change the tax structure in ways
which increased the susceptibility of government income to fluctuations in world markets.
For example, Law 23 of 1897 liberated all exports from taxes except for nitrates and
iodine. If it is assumed that there was less than perfect correlation among fluctuations
in revenues from the different exports, the risk of fluctuations in government revenues
was thereby increased.
26. The estimation of regressions analogous to relations 1.1 and 1.2, but with
ordinary and extraordinary government income as the respective dependent variables,
also did not yield evidence of a significant phase-coincident shift in growth rates.
27. Higher debt ratios were recorded in 1830—35 after the War of Independence,
in 1870—75 after the war with Spain of 1865—66, and starting in 1895 after the Civil
War of 1891. A particularly high debt ratio was not maintained in 1885 after the War
of the Pacific because the acquisition of control over revenues from nitrates made rapid
amortization possible (see column 11). The relatively high ratios in 1830—45, 1870-75,
1900—15, and 1930 also partially reflected foreign borrowing for infrastructure construction.
28. After the War of the Pacific, foreign trade policy was relatively liberal, although
convertibility generally was not maintained and the 1897 tariff law substantially increased
industrial protection.
29. The critical value for a normally distributed series is 2.0; the test statistic is 2.5.
30. The years are 1885, 1886, 1891—94, 1899, 1914, 1919, 1921, and 1922.
31. The test statistic of 0.3 is far below the critical value of 2.0.
32. The low capacity utilization rates in 1914—15, however, may be due to in-
a
42
INTRODUCTION
adequacies in the industrial-sector data used in the underlying series (see Muiioz
[1968:66, 821).
33. For a• satisfactory examination of such interrelationships a general-equilibrium
model is required. See section 2.1 below.
34. Muñoz [1968:13—23] suggests that the post-1860 experience in industrial growth
may have been substantially better than Pinto and others have claimed.
35. The average proportion of government expenditure devoted to development
ranged from 0.26 to 0.47 after 1890 (Table 1.1, col. 9). For a listing of the external
government borrowings and their uses (which included substantial construction of
infrastructure) see Humud [1969:73—80]. Information on investment in human capital
is scarce, but Mamalakis [1971b: 128] does report that literacy increased from 13.5 per
cent in 1854 to 56.1 per cent in 1930 and that compulsory primary education was
introduced (at least de jure) in 1920.
36. Savings also may have been affected positively by the transfer of resources out
of agriculture or by forced savings induced by the inflation (both of which may have
been partly brought on by the international economic regime), but no concrete evidence
is available.
37. Of course marginal, not average, products are of real interest, but to the extent
that the production process can be approximated by a Cobb-Douglas production
function, the marginal and average products are proportional.
38. The following estimates of the sector share of the labor force in Census years
are from Ballesteros and Davis (1963:1761, and may be compared with the estimates
for the sector share of GDP in Table Al, lines 3.1.1.1 to 3.1.1.3; but see also note 20,
above:
Agriculture
Mining
Industry and construction
Transportation
Government
Commerce and services
1907
37.7%
2.8
17.6
3.4
5.3
33.2
1920
36.2%
4.1
15.2
4.8
4.2
35.5
1930
34.7%
5.1
18.3
3.8
7.6
30.5
39. Not only was industry a sector of lower than average labor productivity, but
Muñoz [1968:61] estimates that average industrial labor productivity declined 14.4 per
cent between 1914—16 and 1922—24.
40. I do not imply that the Chilean elite did not act in their own best interests in
this period, but that they did not act in the best national interests of Chile as viewed
decades later by such commentators as Hurtado [1966:701, Lagos [1966:40], Mamalakis
[1971b:133, 488] and Pinto [1962:73—74].
41. The data presented in Ballesteros and Davis [1963] imply this growth rate
(Table A.!, line 5.1). This estimate may be on the low side. Mamalakis [1971b:40]
argues that if the more rapidly growing services sectors are included (Ballesteros and
Davis include only the goods sectors) the rate may be closer to 2.0 per cent. Moreover,
Muñoz [1968:35—40] maintains that for industry the Ballesteros and Davis estimates are
biased downward because of the inclusion of artisan industry in the 1907 base-year
census, but not in subsequent censuses, and because of the assumption of almost no
industrial growth during World War I (owing to a lack of data) instead of the 8—9
per cent annual estimate which Mufloz makes. On the other hand, Muiioz also claims
that the Ballesteros and Davis estimates may be high in the early 1920s because of
insufficient adjustment for depressed international market conditions. Moreover, the
extremely high ratios for imports and exports relative to GDP (lines 1.2.5.7 and 1.2.5.8
AN OVERVIEW
P
•
43
in Table A.l) suggest that the estimates of the level of GDP for the early part of this
period may be low relative to those for foreign trade. Since the foreign trade estimates
are probably more securely based, this last observation implies a higher level of GDP
in the earlier years and a lower rate of growth.
42. Evidence for earlier decades is very limited, but the growth of government
income and expenditures and exports and imports (all in constant dollars) in conjunction
with measures of the importance of related variables relative to total product in 1908—
27 (lines 1.2.6.7, 1.2.6.8, and 3.1.2.3 in Table A.l) are suggestive. For 1844—1927
mean exponential growth rates for ordinary government income and expenditures were
0.043 and 0.044 respectively (with no significant phase-coincident deviations). Govern-
ment income and expenditure may have grown somewhat faster than total product
during this period, but probably not very much so, or the estimate for government current
expenditures relative to GDP fci- 1908—27 (line 3.1.2.3 in Table A.l) would have
been higher. Adjustment for probable population growth implies a growth rate in per
capita product of about the same or slightly higher magnitude as that in the text.
Analogous considerations for exports and imports (including the growth rates in relations
1.1 and 1.2 and the high ratios in lines 1.2.6.7, and 1.2.6.8 in Table A.1) imply a growth
rate in per capita product of about the same or slightly lower magnitude.
•
43. Dc Castro and de Ia Cuadra [1971:11 also refer to the Ballesteros and Davis
r
but ignore the warning that the 1928 and 1929 figures probably are misleading
because of high nitrate prices and the large transitory inflow of foreign capital. Therefore, they emphasize the 2.7 per cent growth rate for 1908—29 instead of the 1.5 per
estimates,
cent growth rate for 1908—27.
44. The 1929 capacity-to-import level, in fact, was not attained again until 1966.
45. See Cohen [1960:8—9], Ellsworth [1945:34—35], J-lirschman [1963:179], Marshall
•
[1957:8—11], and Pinto [1962:109—119] for more detailed discussions of this period.
For data in the text that are not included in Table A.!, see Cohen [1960:8—9] for gold
•
reserves, Hurtado [1966:Tables 13 and 23] for the mining data, Mamalakis [1965:
Appendix Table 1] for the real interest rate, and Ballesteros and Davis [1963:160] for
real government production.
46. For the sources for these and other measures of the decline in the Chilean
economy due to the Great Depression, see Table A.!, lines 1.1.5.1, 1.2.4.1, 1.2.4.2,
1.2.6.3, 1.2.6.7, 1.2.6.8, 2.2, and 5.1 to 5.3; Hurtado [1966:Table 231; and Senado,
•
•
Oficina de Informaciones [1971:21].
47. Moran [1970] presents an interesting discussion of dependencia in Chilean
decisions concerning large-scale copper mining in recent decades.
48. Basic general studies which summarize Chilean economic policies and per-
formance since the Great Depression include Behrman [1974], Cauas [1972], Corbo [19711,
Ellsworth [1945], Ffrench-Davis [1971], GarcIa [1964], Gunder-Frank [1972], Hirschnian
[1963], Instituto de EconomIa [1956, 1963], Lagos [1966], Luders [1968, 1970], Mamalakis [l971b], Marshall [1957], Muñoz [1968], Pinto [1962], Pisciotta [1971], and
Sierra [1969].
•
49. During this quarter century Chile also was headed by a fairly wide range of
governments. One immediate result of the catastrophe of the Great Depression was a
short period of political instability. The violent economic contraction in mid-1931 gave
opponents of what Hirschman [1963: 179] refers to as General Carlos lbauiez del
Caznpo's "thinly veiled military dictatorship" an opportunity to force Ibaflez to resign
on July 26, 1931. Fifteen months of instability followed, including the "100 days of
Socialism" in 1932 under Carlos Davila. In October 1932, Arturo Alessandri was elected
with support from the right to head a new government which succeeded in restoring
P
44
INTRODUCTION
political and econotnic stability. Near the end of 1938 a Popular Front government was
elected under Pedro Aguirre Cerda, who was succeeded (after his death in 1941) by
Juan Antonio RIos. In 1946 Gabriel Gonzalez Videla was elected with considerable
support from the left, but during the six years of his administration, the government
moved substantially to the right. In 1952, Ibaflez again was elected as an independent
candidate who, it was widely hoped, would be a strong man capable of eliminating
internal instability. Ibañez served through the end of this quarter century and through
the liberalization attempt of 1956—58.
50. The availability of national accounts starting in 1940 also entered into the
choice of 1940 as the Start of the second subphase.
51. Unless otherwise noted, the sources for the data in this section may be found
in Tables 8.1 and
52. This was probably the highest decadal mean in at least a century, although
information is very scanty for before 1880 (see note 1, above, and Table A.1, line 2.1).
53. The relatively high mean rate of growth of the constant-dollar value of imports
for 1940—46 (Table A.I, line 1.2.4.1) reflects the high rates of growth in 1940 and 1946
(the latter apparently related to pent-up demand) which offset the substantial declines
(40.5 per cent in 1942) resulting largely from shortages of imported supplies during
the war.
54. Import capacity is somewhat understated because the variable does not adequately reflect the increase in the proportion of each dollar of Chilean copper sales
which was returned to Chile (compare lines 1.1.5.1 and 1.1.5.2 in Table A.!).
55. This ratio equals 1 if there is no exchange control and tends to be correlated
with the degree of exchange control, although the correlation is less than perfect because
of the relative narrowness of the black market in foreign exchange and because of the
impact of illegal activity on supply and demand in this market.
56. The index of export capacity to import, however, rose much less (Chart 1.1)
than the constant-dollar value of exports because the Chilean unit value of imports used
in the former index increased more rapidly than the United States deflator used in the
latter index.
57. After 1946, and continuing to the present, devaluations have been a regular part
of the Chilean experience. They generally did not, however, keep up with the inflation.
See section 3.2 and Table A.7.
58. For further descriptions of economic conditions in these years see Cohen
[1960: 23—24], Ffrench-Davis [1971: 160], and Hirschman [1963:199—2021.
59. Of course, in this period, price increases were substantially below the range of
70—80 per cent that prevailed when the 1956 stabilization effort was started, which may
explain the difference in the degree of preoccupation with price stability between the two
administrations.
60. In a comparative sense, Chilean economic growth had not been particularly high
for some time. According to unpublished 1970 UNCTAD calculations, for the 1950—65
period Chile ranked twenty-ninth among 45 developing countries in the degree of increase
of its per capita GDP. For 1960—65, Chile ranked thirty-eighth among 56 countries.
61. These relative measures indicate a slight decline in the importance of the debt
during 1965—70.
62. In addition to Zorrilla [1970], sources on which this summary of the objectives of
the Allende government is based include Allende [19711, Cauas [1972:46—56], CORFO
[1970—71], Gunder-Frank [1972:16—18], Hobsbaum [1971:23—32], ODEPLAN [197la,
197 lb], Pisciotta [1971:236-2371, and Vdskovic [1971:385—399].
AN OVERVIEW
45
63. The imported component of such investment, however, would be lower than in
the past in order to reduce external dependence.
64. June 1972 unemployment had dropped further, to a level of 3.7 per cent.
65. Before the Chileanization and nationalization policies of the previous phase,
large-scale copper mining had been entirely foreign-owned. Under Frei, however, a substantial share of ownership had been returned to Chile. For details, see subsection 4.2.1,
below.
66. De Castro [1972:45] suggests that actual investment may in fact have declined
more than 7.7 per cent because much of the construction investment was done by the
government and measured at factor cost. It was, therefore, overvalued in real terms
because of the very high rate of increase of wages.
67. On the other hand, investment in human capital may have increased. Minister.
of Economics Vtiskovic also suggested that the changing of the structure of ownership,
which was accelerated in this period, was a prerequisite to sustained long-run Chilean
development.
68. The ranges refer to the alternative estimates given in Banco Central [1973b1.
69. In the first three quarters of 1972, shipments were down 21 per cent in current
dollars and export returns increased only 1.5 per cent in current dollars.
70. To provide some perspective, however, the 1971 and 1972 copper prices still
were higher than those received for any year before 1965 and, according to Cauas
[1972:52], were higher than the price which had been used in 1970 in the Central Bank
in its projections for 1971.
71. In the first eight months of 1972 current-dollar registrations were 17.9 per cent
above the level of the comparable period for 1971.
72. The Allende government inherited $3 billion of foreign debts with amortization
and interest payments of $433 million due in 1971, $409 million due in 1972, $410 million
due in 1973, and $388 million due in 1974. In April 1972 the Paris Club (the United
States and ten other major creditors) recommended the adoption of a new payments
program over an eight.year period, including a two-year grace period on 70 per cent of
the Chilean debts which were due between November 1, 1971, and December 31, 1972.
The creditors promised to study "with good will" the refinancing of those debt payments
due in 1973, but no agreement on further rescheduling had been reached by the end of
the Allende government.
73. Perhaps one should say that exchange control was reintroduced, since substantial
control had existed a century earlier.
74. That changes in measures based on price ratios of final to intermediate products
do not induce resource allocation changes except under very special assumptions has
been well-established for some time on a theoretical level. The present results illustrate
how such measures also may be quite inadequate in empirical applications.
75. The recent reduction of foreign ownership which is described in subsections
4.2.1 and 4.3.2 may have reduced the impact of this effect.
76. Nontraditional exports, the expansion of which has long been desired in order
to reduce risks by diversifying, have been particularly discouraged (section 7.2).
77. Revenues from imports have fallen more than total imports because of the
tendency to favor "essential" imports with relatively low tariff rates when restrictions
have been intensified.
78. In the past, of course, pressures resulting from internal inflation have forced
the government to devalue despite the costs, but then generally an attempt was made
•
again to maintain an overvalued NER.